The next stage in quantum key distribution

The internet has become a fundamental part of our society, allowing us to communicate with people around the world instantaneously, as well as enabling us to share information with colleagues and perform seamless banking transactions. However, the internet is not as secure as we would like. We are becoming increasingly aware of how vulnerable these systems are to people being able to intercept our messages.

Anybody who has physical access to a telecommunications network, whether it be a local area network or part of national telecommunication infrastructure, can eavesdrop on the exchanged information, making encryption protocols a necessity in order to preserve privacy and maintain tight communications security.

One emerging solution to this is quantum key distribution (QKD). Based upon the fundamental laws of quantum physics, this technology encrypts messages and transmits the encryption key required to read it at the photonic level (the smallest measurable level of light) alongside the encrypted message.

One key advantage that QKD has over other new forms of telecommunication systems is that it can use the existing telecommunications infrastructure that is already embedded. While the components at the exchanges may need to be replaced, the actual fibre optic cable is more than adequate for performing QKD, which saves the costs of having to dig up roads and replace millions of kilometres of fibre-optic cable.

This is not a new technology, as there are already several companies that offer QKD technology, such as Quintessence Labs in Australia and Toshiba Research in Europe. However, the technology currently costs £100,000 to implement one link, and is bulky (0.5m³) due to the attenuated laser, controller electronics and detector. As such, only organisations with lots of money and a need for high-end security, such as large banks or government organisations, are currently using it.

Ongoing research into QKD by the University of Geneva, in conjunction with NASA’s Jet Propulsion Laboratory (JPL), focuses on a method called the Ekert Protocol and relies on quantum teleportation to transmit the encryption key between the photon emitters and detectors. Although this does not overcome some of the inherent problems of current QKD technology, in terms of scalable networks, it does allow a slightly longer distance to be spanned in a single link by an individual photon.

Fundamentally, every QKD method has the same problem. A single photon can only go so far before it is likely to be absorbed by the fibre-optic cable. Ideally, it would be sent in a vacuum, but this is obviously not a practical solution. As such, the high cost and multiple links required have so far meant QKD has not been a practical solution, except in unique cases.

However, research conducted by Robert Young of Lancaster University and co-founder of Quantum Base, alongside Iris Choi and Paul D. Townsend, could be crucial in shedding new light on how to make this technology more secure and implementable, through creating what Young calls “the second generation of this technology”.

One of the major problems of implementing QKD has been the exorbitant cost, in conjunction with the inherent limitations in terms of distance. Single photons can typically only travel 100km before they are absorbed by the fibre-optic cable.

The key focus of Quantum Base’s research has been to reduce the size of the components down to a micron cubed and to reduce the cost by a factor of 10⁵. “We want to be able to produce a QKD system that costs £1 instead of £100,000,” says Young.  This reduction in size and costs makes the implementation of QKD systems a more economically feasible solution.

Through the use of quantum filters, Young has also designed a technique of detecting when nefarious third parties are attempting to intercept messages.

The technology operates on the photon’s fluctuating nature. The act of observing or measuring the state of a photon will affect the photon itself. “If I send a vertically polarised photon down an optical fibre from Alice to Bob, and it ends up being a horizontally polarised photon when it gets there, then the only way that could happen is if nefarious parties tried to monitor the particle and the measurement went wrong,” explains Young.

These measurements are performed millions of times over a random selection of the photonic data that is transmitted (as random polarisations). When the receiver’s results are compared with what the sender transmitted, these will highlight any discrepancy in the results. If the discrepancies are higher than 25% (there will be some discrepancies due to the random noise, but these are usually pretty low), then the receiver and sender will be able to determine that a third party has intercepted their communication.

This technique does not, in itself, prevent intrusion by third parties. Instead, detection would allow immediate action to be taken, such as invalidating a financial transaction, or prompt cessation of information transfer.

Through miniaturising QKD system components using standard semi-conductor electronics, Young and his team are seeking to integrate them into the circuits on a chip. “We are tackling all problems independently, so we are working on photonic circuits, light sources and detectors, and then putting them all together to provide the final solution,” explains Young. “Each of the individual components works well, and we expect to have them all together on [a single] chip within 24 months.”

As well as the primary data security element, there is a second benefit to using QKD systems. Not only are they secure from third-party intrusion, but photons use the minimum packet size in terms of energy. “If I can send and detect them efficiently, then it also saves as much energy as possible,” explains Young. “This means you can cram [more] bandwidth down an optic fibre.”

Naturally, there is a vast difference between the ideal conditions found in a laboratory and the real-life conditions found in a typical fibre-optic network. However, Young’s research paper proved that secure quantum communication could be performed in a practical network. Realistic levels of classical data were included during the trial, and they were able to transmit secure quantum key data during the quiet periods between the bursts of noise generated by Raman scattering (the inelastic scattering of a photon) from the conventional data pulses propagating in the system. 

Young is aiming to have the first practical, economically viable and scalable short-term solutions ready for deployment within two years.  He initially foresees small university campuses and corporate headquarters as being the ideal clients. These organisations will often have tens of thousands of computers and other network devices connected to a single network within only a few kilometres of each other, which is well within the limits of QKD.

Looking further ahead, within five to 20 years (the latter being a conservative estimate), he believes QKD will be everywhere. It has already been proven in previous experimentation that QKD and quantum communication can operate on the fibre-optic cables found within our existing telecommunication infrastructure. This means that much of the back-end required for QKD is already in place and will only require component upgrades to incorporate this new technology. Many of these components are upgraded every few years anyway, so this is not as big a problem as it may initially seem.

Young admits it is difficult to predict when QKD will be available for mass roll-out due to the amount of research and development within the field of quantum physics. Quantum teleportation and quantum memories may work in laboratory conditions, but they do not work very well and are not implementable in large volumes at the moment. These are not insurmountable challenges, and it should be a case of ‘when’, rather than ‘if’, they are overcome.

The security solutions offered by QKD systems provide a much-needed lifeline to the telecommunications networks against the innumerable threats these systems face. Now that QKD is being developed into a more cost-effective method, there is no reason why companies wishing to preserve their data security against third-party interlopers should not consider looking at what QKD will have to offer.